Heparan Sulfate Is a Cellular Receptor for Enteric Human Adenoviruses

Figures

• 

  Figure 1. Receptor characterization for the short fiber of HAdV-F40: recombinant fiber knob binding to (A) A549 cells pretreated with the proteases—ficin and proteinase K—which degrade cell surface proteins; (B) A549 cells pretreated with P4—which inhibits glycolipid synthesis; (C) A549 cells pretreated with V. cholerae neuraminidase—which cleaves sialic acid. (D) CHO cells expressing or lacking sialic acid pretreated with V. cholerae neuraminidase (Neu). Pro-5 cells express sialic acid and Lec2 cells lack sialic acid expression. (E) SHSY-5Y cells expressing polysialic acid and SK-N-SH lacking polysialic acid, (F) A549 cells pretreated with heparinase III—which cleaves heparan sulfate. Binding was determined by geometric mean of fluorescence on a FACSLSRII flow cytometer and shown as % binding of fiber knobs on untreated cells in (A–C,F). All experiments were performed three times with duplicate samples in each experiment. Error bars represent mean ± SEM. * P < 0.05 and *** P < 0.001 versus control.

• 

  Figure 2. Effect of sulfation on FK binding: (A) 40SFK binding to A549 cells. Fiber knobs were pretreated with a broad range of concentrations of different glycosaminoglycans (GAGs). (B) Fiber knob binding to CHO cells expressing or lacking heparan sulfate (HS). HS expression levels are shown with an anti-HS antibody. The parental cell line CHO-K1 expresses GAGs and CHO-ΔHS specifically lacks HS but expresses other GAGs. Binding was determined by geometric mean of fluorescence on a FACSLSRII flow cytometer and shown as % binding of fiber knobs on untreated cells in (A). All experiments were performed three times with duplicate samples in each experiment. Error bars represent mean ± SEM. Error bars represent mean ± SEM. *** P < 0.001 versus control.

• 

  Figure 3. Importance of heparan sulfate on infection and transduction. (A) Infection of heparin pretreated HAdVs on A549 cells. The number of infected cells were quantified by immunofluorescence and shown as % infection of untreated HAdVs. (B) Transduction of HAdV-F41 GFP vector on CHO cells expressing or lacking heparan sulfate (HS) and (C) transduction of heparin pretreated HAdV-F41 GFP vector on CHO-K1 cells. The parental cell line CHO-K1 expresses glycosaminoglycans and CHO-ΔHS specifically lacks heparan sulfate but expresses other glycosaminoglycans. The number of infected/transduced cells were quantified by immunofluorescence and shown as % infection of untreated A549 cells. All experiments were performed three times with duplicate samples in each experiment. Error bars represent mean ± SEM. *** P < 0.001 versus control.

• 

  Figure 4. Effect of soluble heparin on virion binding and uptake in A549 cells. (A) 35S-labeled HAdV-F40 and -F41 virion binding to A549 cells. Virions were pretreated with different concentrations of soluble heparin. Binding to cells was quantified as counts per minute (CPM) by a liquid scintillation counter and represented as % binding of untreated virions. (B) HAdV-C5, -F40 and –F41 uptake in A549 cells. The number of intracellular virions were quantified by immunofluorescence and shown as a relative % of internalized virus in the untreated sample. The experiments were performed three times with duplicate samples. Error bars represent mean ± SEM. * P < 0.05 and *** P < 0.001 versus control.

• 

  Figure 5. Effect of pH treatment on receptor usage: 35S-labeled HAdV-F40 and HAdV-F41 virion binding to HAP1 WT cells, HAP1-∆GAG cells and HAP1-∆CAR cells. The parental cell line HAP1 WT expresses glycosaminoglycans (GAGs) and CAR. HAP1-∆GAG lacks expression of GAGs, while HAP1-∆CAR lacks expression of CAR. (A) Untreated (neutral pH) HAdV-F40 or (B) acidic pH treated HAdV-F40 or (C) untreated (neutral pH) HAdV-F41 or (D) acidic pH treated HAdV-F41 were preincubated with heparin and cells were preincubated with 41LFKs before binding. Binding to cells was quantified as counts per minute (CPM) by a liquid scintillation counter. All experiments were performed three times with duplicate samples in each experiment. Error bars represent mean ± SEM. * P < 0.05, ** P < 0.01 and *** P < 0.001 versus control.

• 

  Figure 6. SPR analyses of HAdV-F40 and -F41 interactions with CAR and heparin and 40SFK interaction with heparin: (A) Untreated (neutral pH) HAdV-F40 or (B) acidic pH-treated HAdV-F40 or (C) untreated (neutral pH) HAdV-F41 or (D) acidic pH treated HAdV-F41 binding to immobilized CAR. (E) Untreated (neutral pH) HAdV-F40 or (F) acidic pH-treated HAdV-F40 or (G) untreated (neutral pH) HAdV-F41 or (H) acidic pH treated HAdV-F41 or (I) untreated (neutral pH) 40SFK or (J) acidic pH treated 40SFK binding to immobilized heparin. Interactions are displayed in response units (RU). All experiments were performed three times. Displayed is one representative set from each experiment.

• 

  Figure 7. Effect of GAGs on fiber knob binding to and infection of HuTu80 cells: Binding of (A) heparin or (B) chondroitin sulfate mix or (C) hyaluronic acid pretreated HAdVs on HuTu80 cells. Binding was determined by geometric mean of fluorescence on a FACSLSRII flow cytometer and shown as % binding of untreated fiber knobs. Infection of (D) heparin or (E) chondroitin sulfate mix or (F) hyaluronic acid pretreated virions to HuTu80 cells. The number of infected cells were quantified by immunofluorescence and shown as % infection of untreated HAdVs. All experiments were performed three times with duplicate samples in each experiment.